A formal methodology for developing ring-fencing arrangements and setting conditions.
Howard Lubow (email@example.com) is president and Ryan Pfaff (firstname.lastname@example.org) is a senior manager at Overland Consulting in Leawood, Kansas. J. Robert Malko is a professor of corporate finance at Utah State University. Leslie Moore Romine is chief staff counsel for the Maryland Public Service Commission.
For decades, state regulatory commissions have been searching for techniques to help adequately safeguard public utilities from the increasingly diversified activities of utility holding companies. One of the authors of this article wrote the following over 30 years ago: “The question is not should we have diversification. We have it. It is pervasive throughout the industry. It has been here for a long time and it is going to be here for the foreseeable future. The question is not whether, but rather, how does a regulatory commission deal with the problems posed by diversification?”1
Since the repeal of Public Utility Holding Company Act of 1935 in 2005, merger and acquisition activity has substantially consolidated the number of companies in the U.S. utility industry, and that consolidation naturally has led to more diverse operations by public utility affiliates. This changing landscape has posed challenging questions to regulators about how best to insulate a regulated utility company in its jurisdiction from the business activities of its affiliates. Many state commissions have attempted to address these concerns by imposing additional regulations to protect the utility. This process is generally called “ring-fencing” because these additional measures act as a protective fence that provides some separation between the utility and the potentially risky activities of its affiliates.
The framework for assessing ring-fencing measures described below utilizes operational and financial risk data of the utility holding company to provide regulators with an objective and systematic method for determining what constitutes an adequate degree of ring-fencing.2
Ring Fencing and PUHCA
The relative stability of public utilities during the last 75 years is a striking contrast to the late 1920s and early 1930s when the public utility industry was one of the most corrupt and volatile industries in the country. From 1929 to 1936 there were 53 bankruptcies of utility holding companies.3 The instability of this industry was largely due to the pyramiding structure utilized by industry participants during this time.
To help combat this problem, Congress passed the original PUHCA in 1935 (“1935 PUHCA”). Although 1935 PUHCA proved quite effective in adding stability to the industry and stemming utility bankruptcies, its appropriateness was called into question during the latter part of the 20th century. One argument against the act was that it restricted the activities in which the owner of a utility could participate. This effectively limited the investment available to the utility industry because most companies were not legally authorized to own a utility.
In 1992 Congress amended 1935 PUHCA, and in 2005, after much debate between the different stakeholders, Congress repealed 1935 PUHCA replacing it with the Public Utilities Holding Company Act of 2005 (“2005 PUHCA”) as part of the Energy Policy Act of 2005. While there is still some debate among industry participants regarding whether the repeal of 1935 PUHCA has helped the public utility industry, it has undoubtedly caused an increase in utility merger activity. Recent utility mergers that would have been prohibited under the 1935 act, but allowed after 2005, include the FirstEnergy-Allegheny Energy merger, the Exelon-Constellation Energy merger and, more recently, the Duke-Progress Energy merger.
In addition to the increased merger activity, the repeal of 1935 PUHCA has resulted in an increase in the unregulated operations by utility affiliates. The older act had restricted the operations of a utility holding company to businesses that were “reasonably incidental, or economically necessary or appropriate” to the operations of the integrated system. With the repeal of 1935 PUHCA, this restriction is no longer in place. Public service commissions must now, as part of the merger review process, critically evaluate companies seeking to acquire public utilities within their jurisdictions. They must determine the nature and extent of risk that the acquiring company’s affiliates pose and determine whether additional ring-fencing of the subject utility is needed to protect it from its affiliate companies. Since it’s a public service commission’s responsibility to ensure that the public utilities under its jurisdiction are protected, it might be tempting to assume that these commissions should always impose strict ring-fencing restrictions in order to protect the utility. However, the strictest approach might not always be the most appropriate or efficient alternative.
Pros and Cons of Ring-Fencing
Public service commissions are facing not just increased merger activity, but also the need to consider increased unregulated operations within utility holding companies. Regulators deciding on the level of ring-fencing to implement likely will seek to weigh the benefits and costs.
The benefits of ring-fencing are primarily reducing risk exposures to the utility by isolating the utility operationally and financially from its riskier affiliates. But although isolating the utility can be advantageous from a risk standpoint, it can be detrimental from an economic standpoint. After all, the benefits of economies of scale are enhanced when fixed costs are spread among a larger number of affiliates. A classic example of the benefits of economies of scale occurs when two companies merge and combine their support functions (human resources, accounting, etc.). If ring-fencing measures are so restrictive that they unnecessarily preclude two affiliate companies from combining their operations in an economically beneficial way, ratepayers and other stakeholders will be harmed.
Ring-fencing also can have implications on capital market access. The public utility industry is one of the most capital intensive industries in the country. Depending on its size and infrastructure requirements, a utility might spend hundreds of millions, or even billions of dollars in capital expenditures during a particular year. Given the relatively reliable and predictable cash flow that is assumed in a rate-regulated environment, utilities generally are able to finance much of their external capital needs with debt. These unique characteristics of the utility industry make credit ratings a primary area of concern for most utilities (and, by extension, utility regulators) because a weak credit rating could mean a higher cost of debt and limited financing options.
In the post-1935 PUHCA environment, most utility holding companies have multiple subsidiaries, both regulated and unregulated. Depending on the company’s unique situation, management often chooses to issue debt at multiple entities within the structure, including the parent company. The rating agencies approach to evaluating a subsidiary of a larger holding company is to consider both the creditworthiness of the standalone company and the creditworthiness of the consolidated company. The major rating agencies (Moody’s, Standard & Poor’s, Fitch) generally view the credit rating of a utility subsidiary as constrained by that of its parent, because a financially weak parent could siphon off cash from the utility or, in the extreme, force the utility subsidiary into a consolidated bankruptcy filing.
Rating differentials (sometimes referred to as “notching”) between parent and utility are possible through ring-fencing measures. To justify any significant notching, however, rating agencies require extensive ring-fencing conditions. Standard & Poor’s provides the clearest guidance, stating that it “takes the general position that the rating of an otherwise financially healthy, wholly owned subsidiary is constrained by the rating of its weaker parent… Depending on the ‘stand-alone’ strength of the subsidiary, a package of enhancements … may be effective to raise the rating of the subsidiary a full rating category over the credit quality of the consolidated entity.”4 Even with robust ring-fencing measures, a utility is generally able to achieve a maximum three notch differential from its parent.5
While notching between utility and parent might be a primary consideration to regulators under certain circumstances, notching should more often be viewed as an ancillary issue and not the primary determinant of regulatory action when evaluating ring-fencing measures. The framework described in the following section isn’t a way to manage credit ratings. Credit ratings are based on a combination of qualitative and quantitative factors, and outside attempts to influence the rating agencies’ decisions (unless extreme and robust measures are taken) are likely to be ineffective. More importantly, not all ring-fencing conditions elicit additional notching, but can be nevertheless important considerations for regulators responsible for protecting the operational and financial stability of the utility.
Setting a Framework
To achieve optimal regulation, it’s necessary to balance the two extremes of overly burdensome ring-fencing measures and inadequate ring-fencing measures that could leave a utility vulnerable to the risks of its affiliates. By using an objective matrix of relevant factors, regulators can derive a mix of ring-fencing conditions that are adequately protective, but not overly restrictive.
Two types of risks that are most germane to utility ring-fencing decisions have been considered in the framework provided below: the level and nature of unregulated business activities performed by the utility’s affiliates, and the financial risks of the utility holding company.
The first risk considered in this matrix is the risk related to unregulated business operations. This risk is termed as the “Diversified Operations Risk.” This risk is calculated by first identifying the major unregulated business activities performed by utility holding companies. These activities are then given a “Relative Risk” score from 1 (lowest risk) to 5 (highest risk). In order to incorporate the degree of unregulated business activity, this Relative Risk score is then multiplied by the percentage of holding company revenue represented by this activity. The sum of this score is then added to the “Financial Risk” score to produce the Overland Consulting Matrix score, or “OCM” score.
The Financial Risk score is a function of the parent corporate credit rating. Corporate credit ratings are an ideal proxy for financial risk because ratings agencies formulate these ratings based on a thorough review of the subject company’s financial risks. Furthermore, these ratings are often reviewed and relied upon in the context of regulatory proceedings, so they are familiar to most regulators. Similar to the Diversified Operations Risk score, a numeric Relative Risk score is attributed to all investment grade credit ratings. These scores range from a low risk score of 1 (for strong investment grade credit ratings) to a high risk score of 5 (for weak investment grade credit ratings). As stated in Figure 1, under this framework a utility holding company with a non-investment grade corporate credit rating would immediately trigger the most stringent ring-fencing conditions.
The total OCM score is derived by summing these two risks. The minimum OCM score a holding company could receive is 2 (which assumes a Diversified Operations Risk score and a Financial Risk score of 1). The maximum OCM score a holding company could receive is 10 (which assumes a Diversified Operations Risk score and a Financial Risk score of 5). Once calculated, the total OCM score is placed into one of three categories: Low Risk – Less than 4; Moderate Risk – Between 4 and 7; and High Risk – Greater than 7.
In 2008 Constellation Energy Group (CEG) was an energy holding company headquartered in Baltimore, Md. Within its corporate structure, CEG was the direct parent of Baltimore Gas & Electric (BG&E),6 a large electric and gas utility subsidiary regulated by the Maryland Public Service Commission (MPSC). In addition to its regulated utility business, CEG participated in a variety of energy-related activities through its unregulated subsidiaries.
On Sept. 15, 2008, shares of CEG stock dropped 73 percent based on issues with its energy trading business. Soon after, CEG’s trading counterparties began requiring additional collateral that CEG didn’t have sufficient capital to satisfy. This placed CEG on the brink of insolvency. BG&E, although solvent on a standalone basis, likely would have been consolidated into CEG’s bankruptcy filing.7 CEG ultimately avoided bankruptcy by receiving an injection of capital from MidAmerican Energy and negotiating a sale of a portion of its nuclear assets to the French energy conglomerate Électricité de France (EdF). As part of CEG’s sale to EdF, the MPSC required that CEG implement a robust set of ring-fencing measures to protect BG&E in the future.8 These additional ring-fencing measures greatly enhanced the protection of BG&E – ultimately resulting in an upgrade of the utility’s credit ratings – but the fact that Maryland’s largest utility came dangerously close to being pulled into a bankruptcy proceeding caused by the operations of its unregulated affiliates highlights the need for state commissions to engage in proactive, rather than purely reactive, ring-fencing assessments.9
State commissions can utilize a matrix approach to periodically assess the level of ring-fencing measures appropriate for a holding company whose risk profile may change over time.
Applying the Matrix
To avert catastrophes like what nearly occurred at BG&E in 2008, regulators could periodically assess the business activities of utility holding companies by methodically quantifying its risks – through an approach such as a calculated OCM score.10 This risk score can then be applied to the matrix to determine whether the risk to the subject utility is high, moderate, or low. If the current ring-fencing provisions in place at the utility are less robust than what is implied by the matrix, regulators could consider ordering – or, at a minimum, strongly recommending – more substantial ring-fencing measures.
In the specific case of CEG and BG&E, while the macro events of 2008 were unpredictable, the growing riskiness of CEG was a gradual process. As discussed in CEG’s 2008 10-K Filing: “Over the past few years, our merchant energy business, which includes our trading operations and international commodities operation, grew rapidly. As that business grew, so too did its need for capital, particularly to fund the business’ collateral requirements… The growth of our merchant energy business and its increased need for collateral, coupled with significant volatility in commodity prices in 2008, required us to post substantial amounts of incremental collateral to our counterparties. The asymmetrical nature of the Customer Supply business’ collateral posting requirements compounded the magnitude of the problem, negatively impacting our overall liquidity.”
The gradually changing risk profile of CEG is evident when applied to the matrix. For demonstration purposes, CEG’s OCM score was calculated in 1999 and 2008. In 1999, when CEG was still composed primarily of regulated operations and its credit rating was strong, CEG’s OCM score was 3.7211 (see Figure 1).
In 2008, due to CEG’s low investment grade credit rating and its growing trading operations, as well as the impact of deregulation of electric generation in the state of Maryland, CEG’s OCM score increased to 8.39 (see Figure 2).
Applying CEG’s 1999 OCM Score of 3.72 to the matrix (see Figure 3) would result in a “Low Risk” score. This score indicates that the risk to BG&E from its holding company is not significant and implies the need for only minimum standard ring-fencing measures. Applying CEG’s 2008 OCM Score of 8.39 to the matrix would result in a “High Risk” score. A high risk score indicates that the risks to the utility posed by the activities of its affiliates are substantial and, therefore, the ring-fencing protections should be monitored closely and, if necessary, enhanced.
The logic of the matrix is straightforward: the greater the OCM score, the greater the need for protective measures for the utility. There are some ring-fencing measures that are fundamental, regardless of parent company risk. These include maintaining a separate utility board of directors and maintaining separate utility books and records. This level of ring-fencing would be appropriate in low-risk utility holding companies, as was the case for CEG in 1999. There are other ring-fencing measures that aren’t necessary in low and mid-risk situations, but might be necessary when a utility’s holding company is engaged in particularly risky activities, as was the case for CEG in 2008.
The ring-fencing measures listed in the Figure 3 matrix aren’t meant to be exhaustive. Nor is it recommended that public utility commissions blindly use this matrix without consideration of case specific circumstances. This matrix could be used in a fashion similar to how credit ratings agencies use their credit ratings matrices, which is to say it provides an objective starting point that public utility commissions can then modify based on the specific situation. By using objective criteria upon which to base ring-fencing decisions, regulators can drastically reduce the amount of time and effort that is spent in devising ring-fencing provisions. Employing more objective criteria will also benefit the utilities by decreasing the level of regulatory uncertainty regarding the imposition of ring-fencing measures.
The repeal of 1935 PUHCA has made it possible for utility holding companies to engage in highly diverse operations. It has also made the jobs of regulators more difficult by forcing them to evaluate complex affiliate issues. One of the primary concerns of public utility commissions is whether, and to what extent, public utilities should be shielded from the operations of their affiliates through the use of ring-fencing measures.
A framework providing clear and objective guidance when imposing ring-fencing conditions can greatly reduce the time and resources required for this analysis. The framework provided is subject to further refinement as an analytical tool for regulators. It has been developed as a broad version of the model so that regulators can begin to consider the use of ring-fencing conditions in a more systematic fashion, not only in merger proceedings, but in the context of the ongoing regulatory oversight function.
2. The term “regulators” in this article is primarily meant to include U.S. state public service commissions. However, the basic precepts that were used developing the framework are based on regulatory policy generally applicable to regulated utilities.
3. Lynn Hargis, “PUHCA for Dummies: An Electricity Blackout and Energy Bill Primer,” Public Citizen’s Critical Mass Energy and Environment Program, September 2003. http://www.citizen.org/documents/puhcafordummies.pdf
6. In 2008 BG&E had over 1.2 million electric customers and roughly 650,000 gas customers. Figures obtained from “Electric Operating Statistics” and “Gas Operating Statistics” sections of CEG’s 2008 10-K Filing.
8. These enhanced ring-fencing provisions, which included a non-consolidation opinion and the creation of a bankruptcy-remote special purpose entity, were ordered by the MPSC in Order No. 82986, regarding Case No. 9173, In the Matter of the Current and Future Financial Condition of Baltimore Gas and Electric Company.
9. CEG avoided bankruptcy first through an injection of capital from MidAmerican Energy and then through a sale of corporate assets to EdF. Overland Consulting represented the MPSC Staff in the hearings related to CEG’s asset sale to EdF.
11. In some cases, CEG provided descriptions of its business segments and provided the revenues for these business segments in total, but CEG didn’t provide a breakout of inter-segment revenue. For example, CEG’s 1999 10-K didn’t split merchant generation revenues and wholesale energy trading revenues. For purposes of this illustration, the assumption was made that revenues were split evenly among these different activities.